Alright I know sometime ago that I wouldn't create style loras anymore but this case is slightly different! I was messing around with GPT 4o on the bing image creator and decided to try and see if I could create a "style" out of its realism generation! And of course I am a bit of satisfied and disappointed....Of course I should have made more generations (I used around 100 pics) to make this lora. Also, I am fully aware that what dictates the quality of a lora is the quality of the pics used. So next time, I am not going to rely only on quantidy, but also quality as well! So stay tunned for the V2! Anyway, enjoy!Ps: I can't stress enough for you to use the following lora alongside this one: <a target="_blank" rel="ugc" href="https://civarchive.com/models/1115090/illustrious-realism-enhancer?modelVersionId=1253047">https://civarchive.com/models/1115090/illustrious-realism-enhancer?modelVersionId=1253047</a> Ps2: Yes, the face isn't good, but I will work about it.

GPT - 4o Realistic Style illustrious

Now I can say it is more stable! This time, I applied what I learned these few years with Lora training!

Files with SHA256: d55116c4dd50a51d5cec5bde611bbeda104e9215f27ee18fc09d192baa57e32b